Modules for Velodyne 3D LiDAR Calibration and Data Processing
Camera, LiDAR, Velodyne, calibration, 3D marker, point cloud, road detection, obstacle detection
This software contains several ROS (Robot Operating System) modules for processing data from Velodyne 3D LiDAR, e.g. for road and obstacle detection. One of the modules implements a new pipeline for calibration, i.e. mutual pose and orientation estimation, of the camera and LiDAR sensor using a coarse to fine approach. The calibration benefits from a novel 3D marker which can be robustly detected in both the camera image and the LiDAR scan. The method requires only a single pair of camera-LiDAR frames for estimating large sensors displacement. Consequent refinement step searches for more accurate calibration in a small subspace of calibration parameters.
Distributed under LGPL version 3 open source license.